The diagnosis of ductal carcinoma in situ (DCIS) of the breast has risen dramatically during the last two decades and now accounts for 10-15 percent of all breast cancers diagnosed in the US. Women with DCIS are at substantially increased risk for another breast cancer. However, few predictors of second cancer after DCIS have been identified. Mammographic density, or the extent of the breast occupied by radiologically dense tissue, recently has been established as one of the strongest known risk factors for primary invasive breast cancer, with relative risks as high as 5 or more when comparing the most extensive with the least extensive categories of radiologic density. However, density has not been evaluated as a possible predictor of second breast cancer after either invasive breast cancer or DCIS. Most breast cancers after DCIS are located in the involved, or ipsilateral breast. The extent to which the ipsilateral breast is occupied by radiologically dense tissue at diagnosis may reflect the activity of hormones and other factors capable of stimulating the growth and proliferation of occult malignant cells that are often left behind after breast-conserving surgery for DCIS. The aim of this study is to determine whether mammographic density at diagnosis of DCIS predicts risk of subsequent ipsilateral breast cancer. The study will include participants in the National Surgical Adjuvant Breast and Bowel Program (NSABP) B-17 clinical trial and take advantage of data already collected on treatment, patient and pathologic factors, and subsequent cancer. The B-17 trial population includes 814 women diagnosed with DCIS and randomized to excision alone or excision plus radiotherapy. A recognized expert will assess mammogram from the index DCIS diagnosis for parenchymal pattern and percent of the breast occupied by radiologically dense tissue, without knowledge of second cancer status. The NSABP B-17 trial population provides a unique opportunity to conduct an extremely cost-efficient and well-controlled study of the association between mammographic density and breast cancer risk after DCIS. The results will provide information that may assist patients and their doctors with treatment decisions. In the future, knowledge about mammographic density may also help target those women with DCIS who are most likely to benefit from therapies, such as tamoxifen, that are mediated through a process reflected directly or indirectly by mammographic density.